Blossom: an Anytime Algorithm for Computing Optimal Decision Trees - LAAS - Laboratoire d'Analyse et d'Architecture des Systèmes Access content directly
Conference Papers Year : 2023

Blossom: an Anytime Algorithm for Computing Optimal Decision Trees

Emir Demirović
  • Function : Author
  • PersonId : 1129473
Louis Jean
  • Function : Author
  • PersonId : 1257396

Abstract

We propose a simple algorithm to learn optimal decision trees of bounded depth. This algorithm is essentially an anytime version of the state-ofthe-art dynamic programming approach. It has virtually no overhead compared to heuristic methods and is comparable to the best exact methods to prove optimality on most data sets. Experiments show that whereas existing exact methods hardly scale to deep trees, this algorithm learns trees comparable to standard heuristics without computational overhead, and can significantly improve their accuracy when given more computation time, even for deep trees.

Keywords

Fichier principal
Vignette du fichier
hal.pdf (610.24 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Licence

Dates and versions

hal-04108022 , version 1 (26-05-2023)

Licence

Identifiers

  • HAL Id : hal-04108022 , version 1

Cite

Emir Demirović, Emmanuel Hebrard, Louis Jean. Blossom: an Anytime Algorithm for Computing Optimal Decision Trees. International Conference on Machine Learning, Jul 2023, Honolulu, United States. ⟨hal-04108022⟩
70 View
109 Download

Share

Gmail Mastodon Facebook X LinkedIn More